Whole~cell voltage~clamp and single channel recording techniques were used to study drug interctions with N~methyl~D~aspartate (NMDA) and non~NMDA receptor coupled cation and GABAA receptor~coupled Cl~ channels in cultured hippocampal neurons and with voltage~dependent K+ channels in fibroblasts transfected with K+ channel genes. The aim of this work was to explore new strategies for the rational development of antiepileptic drugs based upon their interaction with neuronal ion channel systems. Work focused in the following areas: (i) characterization of the actions of felbamate on NMDA and GABAA receptors; (ii) interaction of remacemide and its des~glycinated metabolite (FPL[12495) with NMDA receptors; (iii) studies on 2,3~benzodiazepine non~NMDA antagonists (GYKI 52466 analogs); (iv) interaction of a novel scorpion toxin (rogotoxin) with the cloned Kv1.2 K+ channel; (v) interaction of benzopyran K+ channel openers with voltage~dependent K+ channels in cultured hippocampal neurons; (vi) block of NMDA receptors by polyamines; and (vii) neurosteroid modulation of GABAA receptors. In addition, studies were carried out on the interaction of the anticonvulsant carbamazepine with NMDA receptor responses in cultured cerebellar granule cells using the Ca2+~sensitive indicator Fura~2. Fel~bamate, a promising new antiepileptic agent, was found to inhibit NMDA responses and potentiate GABA responses (via a barbiturate~like effect) at clinically relevant concentrations. This novel combina~tion of actions may account for felbamate's unique clinical profile. Remacemide, an antiepileptic undergoing clinical investigation, is des~glycinated in vivo to form 1,2~ diphenyl~2~propylamine (FPL[12495). We observed that this metabolite produces a stereoselective open channel block of NMDA receptors, supporting the view that remacemide may serve as a prodrug for an NMDA antagonist. We have pre~viously demonstrated that the 2,3~benzodiazepine GYKI[52466 is a potent antagonist of non~NMDA (AMPA~kainate)~type glutamate receptor responses in cultured hippocampal neurons. We now show that certain structural modifications of GYKI 52466 at position 3 can enhance blocking potency. The parallel increase in potency of GYKI[53655 in blocking AMPA~kainate receptor currents and in seizure protection provides further evidence that the anticonvulsant activity of GYKI 52466 and its analogs is due to antagonism of AMPA~kainate receptors. Noncompetitive AMPA~kainate antagonists (i.e., GYKI[52466) could offer advantages over competitive antagonists in treating seizures, particularly under conditions where high glutamate levels would render the competitive antagonists relatively ineffective.